In the two-stage model for cutaneous carcinogenesis, a single subtumorigenic exposure to a carcinogen (initiation) and subsequent chronic epidermal hyperplasia of sufficient magnitude (promotion) can induce benign and malignant neoplasms. We have compelling evidence that the target cells in cutaneous carcinogenesis are keratinocyte stem cells having extraordinary persistence, quiescence, and proliferative potential. Moreover, expression of a6 integrin (a6bri) together with low to indistinguishable expression of the transferrin receptor, CD71 (CD71dim), has permitted the physical isolation of keratinocyte stem cells from other proliferating keratinocytes. Furthermore, recent results suggest that the hemopoietic stem cell determinant, CD34, may be useful for the isolation of keratinocyte stem cells from hair follicles. These surprising findings suggest that we should be able to isolate the target cells in two-stage carcinogenesis if we employed a further selection based upon resistance to 5-fluorouracil (FUres) because the target cells are known to be resistant to the killing action of FU. Our principal objective is to identify and to isolate the target cells in 2-stage carcinogenesis. Our hypothesis is that the keratinocyte stem cells are the target cells and that they have the phenotype a6 iCD71dimCD34briFUres. Our specific aims are as follows. 1) To determine the cell surface phenotype of the epidermal stem cells. 2) To determine whether the briCD71dimCD34briFUres keratinocytes are the latent neoplastic cells in two-stage carcinogenesis. 3) To determine in situ the identity of the a6briCD7 1 dimCD34briFUres epidermal keratinocytes in control and initiated skin. The novel approaches detailed in this proposal makes possible a test of the long-standing hypothesis that the epidermal stem cells are the latent neoplastic cells in carcinogen-exposed skin. The identification and physical isolation of these cells opens the door to identifying the genes attacked by the carcinogen, to determining how initiated cells lose control of their growth and differentiation, to identifying additional markers for keratinocyte subpopulations, and to identifying the homologous target cells in human epidermis.